JP2020530896A - System to detect moving objects - Google Patents

Abstract

In a system (100) for detecting a moving object (200), it has:-a radar device for receiving at least one signal reflected by the object (200) under at least one angle. (10); and-Processing device (20) for determining at least one relative velocity between the radar device (10) and the object (200) and at least one angle for each determined relative velocity. Micro-Doppler analysis can be performed by the processor (20) on the signal received from the object (200);-micro-Doppler analysis is performed with reference to a determined angle for the received signal; and-executed. The type of the object (200) can be determined by the micro-Doppler analysis. [Selection diagram] Fig. 3

Description

The present invention relates to a system for detecting a moving object. Furthermore, the present invention relates to a method of detecting a moving object. Furthermore, the present invention relates to computer program products.

A radar system is set up to emit a radar signal and compare the radar signal reflected by hitting an object with the transmitted radar signal. At that time, many methods are known that can collect various information about the object. One well-known embodiment is an FMCW (Frequency Modulated Continuous Wave) radar in which the transmitted radar signal is modulated by a sawtooth function. In that case, the distance of the object from the radar system can be determined with good accuracy. The object angle, which indicates in which direction the object is from the radar sensor, can be obtained by using multiple antennas or by controlling one antenna so that the signal is emitted in a preset direction. be able to.

The Doppler deviation of the reflected radar signal with respect to the transmitted radar signal may suggest the relative velocity of the object to the radar system. Objects that move on their own, such as pedestrians reciprocating their hands and legs, exhibit characteristic, often periodic fluctuations in measurable Doppler frequencies. Such fluctuations can be analyzed to allow the objects to be classified in more detail.

Patent Document 1 proposes to control a radar system in an automobile vehicle so that micro-Doppler analysis can be performed.

Complex modulation using, for example, chirp sequences can be applied to classify objects by, for example, a radar system in an automobile vehicle that is movable in its own right. However, in that case, the processing can be very costly. For example, two-dimensional Fourier analysis of the difference signal between the transmitted signal and the received signal may be required, which makes a high-performance processing device indispensable.

German Patent Application Publication No. 102015109759

An object underlying the present invention is to provide a simple radar-based technique for detecting a moving object.

In the first aspect, the present invention provides a system for detecting a moving object, which system is:
-With a radar device to receive at least one signal reflected by an object under at least one angle;
-Has at least one relative velocity between the radar device and the object and a processing device for determining at least one angle for each determined relative velocity;
-Micro Doppler analysis can be performed by the processor on the signal received from the object;
-Micro Doppler analysis is performed with reference to the determined angle for the received signal;
-The type of object can be determined by the performed micro-Doppler analysis.

In the proposed system, micro-Doppler analysis is performed and the types of objects are classified based on this. A moving pedestrian has different body parts that move at different speeds with respect to the radar device, so that the speed distribution thus generated is characteristic of the pedestrian over time. obtain. For example, the systems proposed for automobiles have the advantage of being able to provide radar-only pedestrian protection or cyclist protection as a result.

In a second aspect, the invention provides a method of detecting a moving object, which method has the following steps:
-At least one signal reflected by an object is received by the radar device under at least one angle;
-At least one relative velocity between the radar device and the object is determined;
-Micro Doppler analysis is performed by the processor on the received signal and micro Doppler analysis is performed with reference to the determined angle for the received signal;
-The type of object is determined by the performed micro-Doppler analysis.

In one preferred embodiment of the system, it is intended that the reception angle can be determined for different relative velocities. Thereby, the micro Doppler analysis can be performed more precisely.

In one preferred embodiment of the system, it is intended that the angle determination can be performed by correlating the received signals. As a result, reliable determination of received signals obtained from various angles is executed.

One preferred embodiment of the system is intended for the determined angles to be utilized for simultaneous microDoppler analysis of multiple objects having relative velocity distributions that overlap each other. As a result, it becomes possible to make different objects distinguishable from each other according to the spatial direction. At this time, for example, it is preferable that a plurality of pedestrians can be distinguished from each other.

Another preferred embodiment of the system is characterized in that the width of the frequency spread of the received signal and the temporal transition of the frequency spread can be determined by the processing apparatus. In this way, there is an advantage that the classification of moving objects is further improved.

Another preferred embodiment of the system is intended for the periodicity of diffusion of Doppler frequencies to be determined by the processor. In this way, for example, the periodic movement of the pedestrian's limbs can be detected.

Another preferred embodiment of the system is characterized in that limiting the angle estimation to a defined narrow frequency / velocity domain is performed. As a result, there is an advantage that the detection capability of the system can be concentrated in the area of interest.

Another preferred embodiment of the system is characterized in that the radar device is configured as a continuous wave radar device. With such a type of radar device, the distinction of received signals can be realized very well.

Another preferred embodiment of the system is further characterized by further having another radar device preferably configured as an FMCW radar device. This makes it possible to make good use of the FMCW radar device for determining the distance and the first relative velocity, and to make good use of the continuous wave radar for the high velocity resolution of the object.

Another preferred embodiment of the system is characterized in that the radar device has at least one transmitting antenna and at least two receiving antennas, respectively, and the receiving antenna can receive the received signal from different receiving directions. In this way, it is possible to reliably determine the angle at which the signal is received.

The disclosed device configuration requirements will be clarified in accordance with the disclosed method configuration requirements, and vice versa. This is in particular the configuration requirements, technical advantages, and embodiments relating to the system for locating objects around the vehicle, and the corresponding configuration requirements, technology relating to the method of locating objects around the vehicle. It means that it becomes clear from the above-mentioned advantages and the embodiment in a similar manner, and vice versa.

Next, the present invention will be described in detail with reference to a plurality of drawings, including further configuration requirements and advantages.

It is a figure which shows the embodiment of the proposed system. It is a figure which shows another embodiment of the proposed system. It is a principle flowchart of the proposed method. It is a graph as an example for explaining the functional form of the preferable development example of the proposed system. It is sectional drawing of the graph of FIG. It is a partial view of the graph of FIG. It is a figure which shows the partial figure of FIG. 6 which made time, frequency, and rasterization for determining the direction or angle of an object.

The idea underlying the present invention is to analyze the spectrum of relative velocity for one or more objects by micro-Doppler analysis with a radar device. In this way, accurate analysis or classification of single and / or multiple objects can be achieved even in complex scenarios with similar distances and velocities but different directions.

This system can allow a combination of the advantages of different radar devices to analyze not only accurate information about the type of movement of an object, but also accurate information about the location of an object and changes in location. In particular, even when a vehicle having a radar device is moving with respect to the surroundings, it is possible to succeed in analyzing accurate velocity information of an object.

The classification of objects can be clearly improved in this way. In particular, the classification of objects as pedestrians / cyclists can be improved and implemented, thereby providing travel assist systems, for example in automobile vehicles, and / or active and / or passive accident prevention devices. Can be improved and controlled. For example, when it is determined that a pedestrian is on a collision course with a car, a signal for warning the driver or pedestrian can be output. In one preferred embodiment, the system can initiate the automatic deceleration of the vehicle.

The processing device is intended to perform micro-Doppler analysis of the signal received by the radar device. Micro-Doppler analysis can determine whether the motion pattern of an object matches a known motion pattern of a pedestrian. Depending on the level of detail of the micro-Doppler analysis performed, there is the advantage that it is even possible to determine what activities the pedestrian is engaged in.

A radar device configured as a continuous wave radar device can be operated by continuous wave operation, for example, with a period of about 15 to about 25 ms, or in another embodiment a period of about 10 to about 15 ms or about 25 to about 30 ms. .. The accuracy of speed determination by Doppler frequency evaluation can be significantly improved thereby.

In this way, the system can be well adapted to the needs of pedestrian detection, for example with a velocity resolution of about 0.1 m / s for a continuous wave signal transmission time of about 20 ms. It can be embodied, which is sufficient to more accurately analyze the typical speed of a pedestrian. A typical pedestrian speed is about 1 m / s for the torso and up to about 4 m / s for forward leg swing, which results in about 10 to 40 frequency bins. On the other hand, for pedestrians crossing in front, there is a clear reduction in relative speed.

In micro-Doppler analysis, the diffusion of the Doppler spectrum is evaluated for moving objects, and stationary objects and moving rigid bodies that do not have diffusion in the Doppler spectrum are ignored.

By micro-Doppler analysis, the difference signal between the transmitted continuous wave radar signal and the continuous wave radar signal reflected by the object can be analyzed with respect to its frequency distribution. At this time, the analysis is preferably performed by the Fourier transform. At this time, the signal energy can be calculated in the preset frequency domain. The frequency distribution can also be analyzed with respect to its temporal transition, thereby distinguishing between walking and running pedestrian movement patterns, for example.

A preferred development of the proposed system is provided with another radar device, which may be configured on the basis of any measurement principle, preferably on the FMCW principle, which usually applies a frequency lamp of a continuous radar signal. May be done. Other embodiments are also possible, for example, a radar device can be utilized in which the individual spatial angles are mechanically or electronically scanned back and forth to determine the object angle.

The signals of the individual FMCW lamps of another radar device are preferably processed separately from each other. Therefore, it is preferable that the FMCW lamp is analyzed by a well-known one-dimensional Fourier transform. This can be a significantly lower computational cost than a two-dimensional Fourier analysis with a chirp sequence. After Fourier analysis, the detected frequency peaks can be combined with each other across different lamps to separate different objects. As a result, both radar devices can be operated alternately, which makes it easier to perform scans in the same frequency domain.

Both radar devices may, as an alternative, be integrated into a single radar device, in which the integrated radar devices operate with different signals one after the other. For example, at one point in time, it can operate with either an FMCW signal or a continuous wave signal. In particular, these operating methods can be activated alternately. Costs can be saved by reducing the radar equipment. A known radar device can be expanded to the above-mentioned system at an expected cost.

It is preferable to focus only on frequencies below the preset limit frequency, which is determined on the basis of the speed of the radar device relative to the surroundings. As a result, it is preferable to focus only on the signal components assigned to the object approaching the radar device at a higher speed than the radar device moving with respect to the surroundings, that is, the object itself moving with respect to the surroundings. .. Correspondingly, the Doppler frequency of such an object is lower (or numerically higher) than the Doppler frequency corresponding to the negative proper velocity.

FIG. 1 shows a basic aspect of the proposed system. You can see the radar device 10 functionally connected to the processing device 20. The radar device 10 emits a transmit signal that hits an object 200 (eg, a pedestrian) and is at least partially reflected and received as a receive signal under a variety of very similar angles. The processing device 20 performs micro-Doppler analysis using the received signal, from which the types of objects 200 are classified.

The proposed automotive system has the advantage that it can be used as radar-based pedestrian protection. Alternatively, radar-based applications, for example in stationary surveillance systems in the military field, are also conceivable.

FIG. 2 shows a useful application as an example of the preferred developments listed above the proposed system 100 for vehicle 50, which includes a radar device 10, another radar device 30, and a processing device 20. Shown. Radar devices 10 and 30, respectively, have at least one transmitting antenna and at least two, preferably four receiving antennas (not shown), respectively, thereby spatially by at least two receiving antennas. Received signals can be received from a variety of different directions, and these received signals are continuously correlated so that directional information can be derived for the received signal. Both radar devices 10 and 30 may be integrated and configured as one radar device, and in such a case, alternating operation as the first radar device 10 and another radar device is preferable. Around 210 around the car 50 is a moving object 200 represented by a pedestrian in the example of FIG.

It is intended that the system 100 scans the object 200 with radar signals to determine the position information, motion information, and classification information of the object 200. The determined information can be provided for secondary use by an interface 40 which may be configured as a warning device and / or a control device (not shown) in the vehicle of the vehicle 50.

The movable object 200 can move with respect to the surrounding 210. Furthermore, the object 200 may move as such or perform micro-motion. At this time, each part of the movable object 200 (arms and legs in the case of a pedestrian) may move with respect to the surrounding 210 at a speed different from that of the object 200. In such cases, radar devices 10 and 30 can measure not just one Doppler frequency, but the entire span of the Doppler frequency.

For example, a movable object 200 configured as a pedestrian can move relative to its surroundings 210 at a speed of about 5 km / h. Based on the periodic movements of the pedestrian's legs (and often both arms), their Doppler frequency diffusion also fluctuates in a cyclical manner. Maximum speed is given by the torso when both legs are on the ground. This speed decreases along the legs and reaches zero at the feet. Therefore, any Doppler frequency corresponding to the velocity between zero and the velocity of the fuselage can potentially be measured. This is also the point of minimum Doppler frequency diffusion. The foot, on the other hand, reaches about three to four times the torso velocity when swung forward.

The Doppler frequency or the region of the frequency bin determined in this way can be used to correlate the received signals of all the receiving antennas. In this way, so-called "multipurpose estimation" can be embodied, and a plurality of objects arranged under different angles are determined by a single frequency bin.
The FMCW signal, known as itself, is the distance and / or rough motion of the object 200 in order to determine the velocity spectrum of the object 200 sufficiently accurately without the need for complex modulation or evaluation of the radar signal. It is proposed that the determination is made by the first radar device 20 that utilizes. In order to obtain high velocity resolution of the object 200, the micro-motion of the object 200 is additionally determined and analyzed by the radar device 10, preferably using micro-Doppler analysis. At this time, it is preferable that the radar device 10 uses a continuous wave signal (“CW lamp”), that is, does not modulate the transmitted radar signal over time. The determination of the continuous wave signal can be performed as defined longer than a normal ramp of the FMCW method in order to achieve sufficient velocity resolution for the object 200, for example lasting for about 20 ms.

For each frequency bin, the correlation of the received signal can be performed with or without dependence on the output received therein. In this way, the output increase can be detected, or the individual received signals can be correlated, but in the latter case, the calculation cost is high.

For continuous wave signals, only the Doppler effect affects the received signal. On the other hand, the distance of the object 200 plays no role. The difference frequency and the accompanying Doppler frequency directly correspond to the physical speed of the object 200 in relation to the vehicle 50. Since the distance cannot be determined for the continuous wave signal, each scene must be divided into individual objects 200 by the conventional FMCW method. However, since both radar devices 10 and 30 can determine the velocity and angle of the object 200 with respect to the radar devices 10 and 30, it is often possible to uniquely assign the micro-Doppler effect to the detected object 200. It is possible.

Finally, in the basic form of the proposed system, the continuous wave signal can be analyzed almost completely separately from the traditional FMCW lamp.

FIG. 3 shows a flowchart 300 of a method in which another radar device 30 is also used to determine information about the movable object 200, which information is particularly the location or motion of the object 200, and the frequency distribution of micromotion. Is.

In step 305, the object 200 is scanned by another radar device 30, preferably based on the FMCW signal. Other radar methods are also possible as alternatives. The transmitted and reflected signal is qualitatively suggested in the time graph above step 305. This decision is well known in radar engineering and can be carried out by any well known method. As a result of the scan, it is preferable that the first distance d (t) with another radar device 30 and the first relative velocity v1 (t) between the object 200 and another radar device 30 are determined. ..

In step 310, which can be performed alternately with step 305, the radar device 10 scans the object 200 based on a radar signal (continuous wave signal) having the same frequency. The graph suggested above step 310 illustrates the transmitted and reflected signal. As a result of this scan, it is preferable that the second relative velocity v2 (t) between the object 200 and the radar device 10 is determined. At this time, the second relative velocity preferably has a very high resolution, which enables efficient execution of the micro-Doppler analysis.

In step 315, the information determined in steps 305 and 310 is assigned to each other. The first and second information, each containing the same angle, and more preferably the same temporal trend at each angle, targets the same object 200 and can be assigned to each other. Step 315 preferably provides a distance d (t), a velocity v (t), and an angle φ (t) as a combination of first and second information.

In step 320, the frequency distribution of the second relative velocity can be analyzed to determine if the obtained pattern suggests a pedestrian.

For this purpose, the diffusion of relative velocities or Doppler frequencies representing relative velocities are determined and analyzed. In the presence of widespread diffusion, the object 200 is classified as a pedestrian by temporal analysis, in which case the corresponding pattern or the characteristics of such a pattern may be pre-determined and incorporated for comparison. Can be done.

In analyzing the received output, the individual received outputs of all the receiving antennas can simply be added together (“non-coherent integration”), or, as an alternative, one frequency at a reasonable angle. It is also possible to try how well one or more objects can be determined in a bin with sufficiently high quality. For each frequency bin, it is sufficient to be able to determine the angle of the object with high output intensity (based on the strong output difference of the received signal), or simply to save the high computational cost for multipurpose estimation. In addition, it is intended to determine only one angle.

The processing of the continuous wave signal of the radar device 10 is basically the same as the processing of the FMCW lamp of another radar device 30, for example. Following non-coherent integration across all receiving channels, spectral analysis is preferably performed by a fast Fourier transform. At this time, the signal is decomposed into each frequency formed by combining them. The output of each frequency component is then determined in each frequency bin, where the frequency bin corresponds to each defined frequency interval in the overall spectrum.

However, unlike FMCW lamps, frequency peaks are not detected (and are not assigned to each other) here. Each frequency bin with an output above the noise threshold directly suggests the presence of a physical object 200 (radially) with a corresponding velocity. This, of course, occurs for the entire frequency spectrum for the object 200 with the microDoppler effect. The angle estimation is virtually the same as for the FMCW lamp. Similarly, detection of individual frequency peaks is simply unnecessary. In addition to this, there is only a single continuous wave signal, from which the angle can be determined, thus eliminating the need to calculate the angle for each lamp. However, in the presence of micro-Doppler, individual frequency bins replace different lamps.

In the automotive field, the self-motion of radar device 10 can make it difficult to perform micro-Doppler analysis to detect pedestrians. That is, to the moving radar device 10, it seems as if the stationary object 200 is approaching from immediately in front at its own speed. When offset laterally, such apparent speed is reduced by the cosine of the observation angle. At the moment of passing by (ie, at 90 °), the object 200 appears to be temporarily stationary and then moves backwards away from the radar device 10. Therefore, the reflected output of a stationary object 200 is limited to frequencies that correspond to velocities between zero and negative self-velocity with respect to the spectrum. The self-speed refers to the speed of the automobile 50 with respect to the surrounding 210.

Such a relationship is shown in Graph 400 in FIG. Is plotted the time in the horizontal direction t, velocity v in the vertical direction is plotted in dependence on the Doppler frequency f _{doppler (t).} The basic signal 405 represents an object that moves at a speed lower than its own speed, which is relatively negative with respect to the radar device 10, and is therefore considered stationary. Each peak (English peaks) 410 corresponds to an object 200 in the form of a pedestrian. At this time, each peak 410 represents the maximum relative speed generated relative to the second radar device 30 by the walking of the pedestrian.

The transition 420 represents the oncoming traffic in which the automobile 50 is stopped. The border of region 405 represents the negative self-speed- _vego of the vehicle 50.

All other frequencies outside this region are undisturbed by stationary objects. In contrast, in other FMCW lamps, the background clutter is clearly distributed over a wide frequency domain.

Crossing pedestrians are of special importance for pedestrian or bicycle protection in the field of driver assistance. Compared to pedestrians coming from the front, the radial component of their motion is clearly lower in the direction of radar device 10, but not zero. Even when a pedestrian vertically crosses the road on which the vehicle 50 is moving, the pedestrian does not move perpendicular to the radar device 10. Nonetheless, for pedestrians crossing, typically only the relative speed of the legs swung forward is higher than that of an object that is stopped immediately ahead in the direction of travel.

Therefore, only the corresponding frequency components are spectrally analyzed without hindrance. Due to the slow but active movement of the pedestrian in the direction of the radar device 10, the micro-Doppler effect to be analyzed corresponds to the frequency domain just below the Doppler frequency corresponding to the negative self-velocity. .. For self-speed, there is usually a high quality estimate inside the car 50. Therefore, the region of the frequency spectrum associated with the pedestrian can be directly selected.

At the corner, the individual points of the vehicle 50 have different velocities based on the rotational motion. The self-speed of the vehicle 50 is usually determined with respect to the vehicle rear axle. The yaw rate of the vehicle 50, which is also usually known, allows the corresponding speed of the radar device 10 assembled on the front side to be easily derived from it.

Since pedestrians approach the roadway from the side, the measurable speed also decreases due to the lateral offset of the radar devices 10 and 30 with respect to the direction of motion. Pedestrians experience the same apparent speed reduction as an object 200 that is stationary under the same viewing angle. On the other hand, if the movement direction of the pedestrian is the same, the radial component of the original pedestrian movement increases as the observation angle increases.

For the original analysis of Micro Doppler, a method similar in principle to a stationary radar system with a constant transmission frequency is suitable. However, based on the masking of most of the micro-Doppler diffusion, the intensity of the micro-Doppler output, the width of the unmasked diffusion, the amplitude of such width variation over time, and the temporal interval between the two maximum diffusions. / Cycle (and the pedestrian's measured step frequency associated with it) is the main criterion.

FIG. 5 shows the graph of FIG. 4 along the cross section at time point t = τ. In the area of peak 410 on the left side, a wide spread of frequencies of peak 410 of pedestrians can be seen. The peak 410 is generated by the high relative speed of the pedestrian's forward swinging leg and the radar device generated by it. A peak 430 caused by a stopped vehicle 50, which has a low relative speed similar to that of a pedestrian by being stopped, is also observed. Here the receive output for the chassis is rising strongly because it is made of metal. In this way, the pedestrian can be satisfactorily identified, the object 200 as a pedestrian can be classified, and the information can be subsequently processed.

FIG. 6 shows a portion B of FIG. 5 in which time / frequency / rasterization has been performed.

In FIG. 7, the region B in the figure is shown in FIG. A) without rasterization, and the time, frequency, and rasterization of region B are shown in FIG. B), and the frequencies for all measurement cycles are shown in the horizontal direction. Displayed, vertically showing all frequency bins for a single measurement period. Here, the square fields B1, B2, B3, and B4 of time, frequency, and rasterization correspond to frequency bins in the discrete domain or in the defined frequency interval of the analog domain.

No analysis is performed on frequency bin B1. This is because, in relation to the radar devices 10 and 30, basically only stationary objects are expected (or the reception output of stationary objects is expected to be dominant).

In frequency bin B2, each received output is correlated so that the object is obtained from it under one angle, where the object is relative to radar devices 10 and 30 in the form of a pedestrian. Is arranged as a target.

In frequency bin B3, each received output is correlated so that the object is obtained from it under one angle, where the object 200 is relative to radar devices 10 and 30 in the form of a vehicle. Is arranged as a target.

In the frequency bin B4, the object 200 cannot be detected based on the correlation of each received signal.

An additional note in this analysis is that pedestrians, by definition, have stationary parts (stationary legs) and the maximum power output is given by the torso. Correspondingly, there is no omission (without signal output) between the Doppler frequency belonging to the negative self-velocity and the pedestrian micro-Doppler spread. Correspondingly, the classification of an object as a pedestrian is denied even when the maximum spectrum value is significantly deviated from the Doppler frequency belonging to such a negative self-velocity.

The method has the advantage that it can be implemented as software traveling on radar devices 10 and 30, thereby supporting the easy modification of the method.

The proposed system has the advantage that the effects of raindrops need not be taken into account because the output reflected by the raindrops often overlaps with the pedestrian's micro-Doppler effect. Since this is a spatially dispersed event, it has been shown that the angle of incidence cannot often be determined, despite partially sufficiently significant output. Therefore, rain substantially only lowers the signal-to-noise ratio, and in particular, the overall width of the pedestrian's power spread can be determined without hindrance.

10 Radar device 20 Processing device 30 Radar device 100 System 200 Object

Claims (12)

In the system (100) for detecting a moving object (200),
A radar device (10) for receiving at least one signal reflected by the object (200) under at least one angle.
It has at least one relative velocity between the radar device (10) and the object (200) and a processing device (20) for determining at least one angle for each determined relative velocity.
Micro-Doppler analysis can be performed on the signal received from the object (200) by the processing apparatus (20).
Micro Doppler analysis is performed with reference to the determined angle for the received signal,
A system capable of determining the type of the object (200) by the performed micro-Doppler analysis. The system (100) according to claim 1, wherein the reception angle can be determined for different relative velocities. The system (100) according to claim 1 or 2, wherein the angle determination can be performed by correlating the received signals. The invention according to any one of claims 1 to 3, wherein the determined angle is used for simultaneous micro-Doppler analysis of a plurality of objects (200) having relative velocity distributions that overlap each other. System (100). The system according to any one of claims 1 to 4, wherein the width of frequency diffusion of the received signal and the temporal transition of frequency diffusion can be determined by the processing apparatus (20). 100). The system (100) according to claim 5, wherein the periodicity of diffusion of the Doppler frequency is determined by the processing apparatus (20). The system (100) according to any one of claims 1 to 6, wherein the limitation of the angle estimation to the defined narrow frequency domain / velocity domain is performed. The system (100) according to any one of claims 1 to 7, wherein the radar device (10) is configured as a continuous wave radar device. The system (100) according to any one of claims 1 to 8, further comprising a second radar device (30) preferably configured as an FMCW radar device. The system according to claim 9, wherein the radar devices (10, 30) each have at least one transmitting antenna and at least two receiving antennas, and the receiving antennas can receive received signals from different receiving directions. (100). In the method (200) of detecting a moving object (200), each of the following steps is included.
At least one signal reflected by the object (200) is received by the radar device (10) under at least one angle.
At least one relative velocity between the radar device (10) and the object (200) is determined.
A micro-Doppler analysis is performed on the received signal by said processing apparatus (20), and a micro-Doppler analysis is performed with reference to a determined angle for the received signal.
A method for determining the type of the object (200) by an performed micro-Doppler analysis. Provided is a program code means for carrying out the method of claim 11 when progressing on a system (100) for detecting a moving object (200) or when stored in a computer-readable data medium. Computer program products that you are using.

Images